In vivo gene editing
Also known as In-vivo genome editing, Systemic CRISPR therapy, In-body gene editing, DNA base editing
In vivo gene editing delivers editing machinery into the body so that it changes DNA inside a person's own cells. A phase 3 trial of an investigational in vivo CRISPR treatment and a phase 1 base-editing study now report clinical results. The approach has so far been tested on only a few gene targets, and the phase 1 base-editing study delivers its editor to the liver.
In plain language
Gene editing has two hard problems: making the intended change to the DNA, and getting the editing machinery into the right cells. CRISPR tools use a guide molecule to lead an enzyme to a chosen DNA sequence. Some enzymes cut the DNA, while base editors change one DNA letter into another without making the same kind of cut. In vivo treatments deliver these tools into the body itself. Ex vivo treatments take cells out, edit them in a laboratory and return them.
For some liver treatments, the editing instructions are packed into lipid nanoparticles, which are tiny fat-like particles that protect RNA and carry it into cells. After a single dose, the liver takes up many of these particles. The cell makes the editor for a short time, and the change to its DNA can remain after the editor has cleared. So far, the approach has been tested mainly where the target protein is made in the liver. It does not yet reach muscle, brain or blood-forming cells in a dependable way.
Technical explanation
In vivo editing combines a programmable editor with a delivery vehicle. CRISPR-Cas nucleases make double-strand breaks that cells repair, often by disabling the target gene. Base editors fuse a catalytically impaired Cas enzyme to a deaminase so that a single base is changed without a double-strand break. Both need a guide RNA that matches the target sequence, so each therapy has its own guide, editor, carrier formulation and dose.
The clinical evidence reviewed here comes from press and news reports of two phase 3 and phase 1 studies, together with NEJM papers that could not be read in full. In the phase 3 HAELO trial, Intellia reported that a single dose of lonvoguran ziclumeran, an investigational in vivo CRISPR treatment that inactivates the kallikrein B1 (KLKB1) gene, reduced mean monthly hereditary angioedema attacks by 87% compared with placebo over weeks 5 to 28. The same trial is described in an NEJM paper by Cohn and colleagues. In the phase 1 VERVE-102 study, Vafai and colleagues describe an adenine base-editor protein, delivered as messenger RNA in a lipid nanoparticle with an N-acetylgalactosamine (GalNAc) ligand, with a guide RNA that targets PCSK9 in the liver. At the highest of six dose levels, PCSK9 protein fell by 88% and LDL cholesterol by 62%, according to a news report of the interim analysis. A 2025 NEJM report by Musunuru and colleagues describes patient-specific in vivo gene editing for a rare genetic disease. Its clinical content was not checked for this record.
The central technical questions are safety and specificity. Off-target edits at similar sequences, large deletions or rearrangements near the target, and immune responses to the editor or carrier all need sensitive measurement over time. Edits in long-lived cells may persist, so follow-up has to be long enough to detect delayed effects. Delivery to other tissues remains the main barrier to wider use.
What the evidence says
Each statement is typed by the strength of its evidence and linked to the source that supports it.
Reported, unverified Stated by a company or team but not independently confirmed.
In the phase 3 HAELO trial, a single dose of lonvoguran ziclumeran, an investigational in vivo CRISPR treatment that inactivates the kallikrein B1 (KLKB1) gene, reduced mean monthly hereditary angioedema attacks by 87% compared with placebo over weeks 5 to 28, according to the company.
Sources: [1], [2] · Press release, efficacy results (weeks 5 to 28) · medium confidenceThe HAELO trial enrolled 80 people, with 52 receiving lonvoguran ziclumeran and 28 receiving placebo. In the treated group, 62% were attack-free and therapy-free over the six-month efficacy period, compared with 11% on placebo, according to the company.
Sources: [2], [1] · Press release, efficacy results · medium confidenceThe company reports that all treatment-emergent adverse events in the HAELO trial were mild or moderate and that no serious adverse events were observed in the treated arm. Infusion-related reactions, headache and fatigue were among the more common events in that arm.
Sources: [2] · Press release, safety section · medium confidenceIn the phase 1 VERVE-102 study, an adenine base editor targeting PCSK9, given at the highest of six dose levels, reduced PCSK9 protein by 88% and LDL cholesterol by 62%, an absolute fall of 78 mg/dL, in an interim analysis of 35 patients who had completed at least 28 days of follow-up.
Sources: [3], [4] · TCTMD report, highest-dose cohort results · medium confidenceIn the VERVE-102 interim analysis, 15 participants had one-year data, and the PCSK9 and LDL cholesterol reductions were maintained at that point. No figures for the one-year results were reported in the source.
Sources: [4], [3] · TCTMD report, one-year follow-up · medium confidenceIn the VERVE-102 study, 20% of participants had grade 1 or 2 infusion-related reactions and transient elevations in alanine aminotransferase were reported. The treatment was described as generally well tolerated. One participant had a grade 3 aspiration pneumonitis, which the investigators judged unrelated to VERVE-102.
Sources: [4], [3] · TCTMD report, safety section · medium confidenceThe VERVE-102 editor is an adenine base-editor protein delivered as messenger RNA in a lipid nanoparticle containing an N-acetylgalactosamine (GalNAc) ligand, and its guide RNA targets PCSK9 in the liver.
Sources: [4] · TCTMD report, formulation description · medium confidenceLimitations and open problems
- Follow-up is measured in months for most patients. The longest VERVE-102 follow-up reported is one year, for 15 participants, so safety and durability over decades are not established.
- Human data reviewed here come from a small number of gene targets. The VERVE-102 study delivers its editor to the liver, and reaching muscle, brain, lung or blood-forming cells needs different delivery methods that are not yet shown in humans.
- Biomarker changes such as lower PCSK9 or LDL cholesterol show biological activity but do not show fewer clinical events. The phase 1 VERVE-102 study did not test cardiovascular outcomes.
- Off-target and structural DNA changes need sensitive measurement, and laboratory assays cannot rule out every consequence in every tissue over a lifetime.
- A patient-specific editor needs rapid diagnosis, design, manufacture and validation for each person. Nothing in the sources reviewed shows that such treatment can be offered routinely.
- Sterile manufacturing, specialist monitoring, cost and regulatory pathways for one-time treatments that may be permanent are unresolved.
Possible future uses
Ideas, not predictions. None of these has been demonstrated.
- Treatment of organs that are hard to reach surgically, if delivery to those tissues can be made safe and efficient.
- Customised editors for rare diseases with a known causal mutation, if manufacturing and regulatory routes become workable.
- Fewer repeat drug doses for diseases driven by a protein that circulates in the blood and is made in the liver.
Timeline
- 12 Jun 2025NEJM publishes a brief report, 'Patient-Specific In Vivo Gene Editing to Treat a Rare Genetic Disease' (volume 392, issue 22). Sources: [5]
- 13 Jun 2026Intellia Therapeutics announces additional positive phase 3 HAELO results for lonvoguran ziclumeran, an investigational in vivo CRISPR treatment for hereditary angioedema. Sources: [2]
- 13 Aug 2026NEJM publishes the phase 1 VERVE-102 base-editing study for PCSK9 (volume 395, issue 7). Sources: [3]
- 15 Oct 2026NEJM publishes the phase 3 HAELO report for lonvoguran ziclumeran (volume 395, issue 15). Sources: [1]
Futures this could help bring about
Glossary
- In vivo
- In a living organism. For gene editing, the editing components are delivered into the body to reach cells there.
- Base editor
- An enzyme system that changes one DNA letter into another without making a conventional double-strand break.
- Lipid nanoparticle
- A small fat-like particle used to carry RNA or other molecules into cells.
- Guide RNA
- A short RNA sequence that helps direct a CRISPR editor to its intended DNA target.
Sources
- peer reviewed paper
Lonvoguran Ziclumeran: In Vivo CRISPR Gene Editing in Hereditary Angioedema
Danny M. Cohn, Padmalal Gurugama, Hilary J. Longhurst. New England Journal of Medicine, 15 Oct 2026. DOI: 10.1056/NEJMoa2600931. Accessed 21 Oct 2026. - company announcement
Intellia Therapeutics Reports Additional Positive Phase 3 Results for Lonvoguran Ziclumeran (lonvo-z) in Patients with Hereditary Angioedema
GlobeNewswire (Intellia Therapeutics), 13 Jun 2026. Accessed 21 Oct 2026. - peer reviewed paper
In Vivo Base Editing of PCSK9 with VERVE-102 for Hypercholesterolemia
Scott B. Vafai, Jörg Täubel, Thomas Ashdown. New England Journal of Medicine, 13 Aug 2026. DOI: 10.1056/NEJMoa2601283. Accessed 21 Oct 2026. - news
Gene-Editing Therapy Safely Lowers PCSK9, LDL Cholesterol in Phase I: Heart-2
Todd Neale. TCTMD, 1 Jun 2026. Accessed 21 Oct 2026. - peer reviewed paper
Patient-Specific In Vivo Gene Editing to Treat a Rare Genetic Disease
Kiran Musunuru, Sarah A. Grandinette, Xiao Wang. New England Journal of Medicine, 12 Jun 2025. DOI: 10.1056/NEJMoa2504747. Accessed 21 Oct 2026.
Revision history
3 changes since this profile was created.